The current best hope for the successful treatment of pancreatic cancer is the removal of pre-invasive lesions before they become malignant. The early detection of a portion of pancreatic cancer precursors- cystic neoplasms of the pancreas-is possible through high-resolution abdominal imaging. The increased use of abdominal imaging has led to a higher rate of identifying pancreatic cysts, but currently it is not possible to accurately determine which cysts have high malignant potential and should be removed. New biomarkers that could assist that determination would lead to more successful outcomes for patients with pancreatic cysts. Therefore, the goal of this research is to develop biomarkers for distinguishing pancreatic cysts with high malignant potential from those with low malignant potential. Our overall hypothesis is that the expression and glycosylation of specific proteins are significantly different between cysts with high malignant potential and cysts with low malignant potential, and that these molecules form accurate biomarkers for the diagnosis of pancreatic cysts. We have a comprehensive strategy for biomarker discovery, refinement, and validation. Multiple clinical sites will contribute the medical expertise and the high-quality sample sets, and experienced statistical expertise in biomarker research will guide the experimental design for biomarker discovery, pre-validation, and validation. Our technological strategy is built on the powerful combination of novel glycoproteomics biomarker discovery methods and complementary antibody array methods for the high-throughput and precise profiling of multiple protein and glycan candidates. Through the iterative characterization and testing of biomarker isoforms and glycoforms, the performance of the best candidate biomarkers will be refined and improved. Pivotal double-blind validation studies will provide accurate assessments of biomarker performance. The success of this project will result in biomarkers to be validated in clinical settings; high-quality sample sets to be used in ongoing EDRN-associated discovery and validation studies; and an improved understanding of the molecular alterations associated with pancreatic cystic neoplasms.